The Situation:

They wanted you to provide:

Market analyse

First and foremost, I will make a quickly analyse of brazilian market that is going to help to choose a sample data which will be better to describe the whole market. Besides, gain of computer performance to test and valuation models. In the firs moment we just have done a data mining to preprer the data for forecast analyse, and standarize few variables.
Remark about how I create group of Avarage Pack Size, I used a boxplot to set the size of each group:

  • Less than First Quartil and greater than minimum Avarage Pack Size = VERY SMALL
  • Greater than First Quartil and less than Median = SMALL
  • Greater than Median and less than Second Quartil = BIG
  • Greater than Second Quartil and less than maximium Avarage Pack Size = VERY BIG

The first graphic bellow ,volume and value normalized, helps us to undestand the trully moviment of market. In 2016, the volume market sharply drecrease more than the value market in the first semester creating a intersection point, after that, volume drecrease less than before June 2016 as you can note the blue line slope. One of my doubt is after June 2016, problably the price should have impacted the volume, as red line is almost flat. Also we can conclude that has been constant the price increase.

A quick understand in the market by flavor , which one has more impact in the market? Easly notice in the graphic bellow the Milk Chocolate has a huge impact in the total result. Futurely, will be select top3 flavors for dig deeply in analyses.

Without data normalized, still possible to verify the same flavor predominance

Analysing the market by material market, the plastic material has been dominante. Even though, the campaing against plastic is sharply rise and the customer is asking change. The enviromment question need to be a topic to consider in all strategy.

The sugar do not seems be affected along years, the majority consumers rather sugar than diet/light in caloric content aspect.

The big package leader has a interest movement in the graph below, in 2015 the value market had a fall back without rise the volume. In other words, is possible to say the demand is inelastic. But in the last months, the demand is elastic, so unless you considers macro economics facts, is possible to say the market reached the maximium demand for this size, also reached the right price. The very small package, is a profitable size, future analyses will be consider these two type of package

To sum up, the table below show by numbers the value market, volume market and price market by month and year.

There are 2 more variables that could have strong influence in the market oscilation, Coverage and Shelf Life. Those charts below describing the mean by month and year. As you can notice, the coverage is decreasing since January 2015 so would be a strong potencial for be one of reason of volume market result. The shelf life, in overall, keep in the same range and is clear the sazonality in the end of the year. The histogram and density chart help us to detect any outlier which could mess up the mean result.

A good way to confirm the variable relation is use correlation matrix. The more is red, more is inverse correlated the variables. In the same way of blue color, the more blue is, more is correlect in the same direction. Can be conclued:

  • The price variable is inverse correlect to Volume, Coverage and Shelf Availability
  • The volume variable is correlect to Coverage and Shelf Availability

The graph above, can be understanded 3 diferent things:

  • normal distribution in scatter plot and red line, more the red line is in diagonal, more is equally distributed.
  • in the diagonal, histogram and densisty red line.
  • Correlation result.

Help to confirm the correlation matrix result.

Second part

The second part is use all the information that was shown then filter the data to apply forecast model. The goal is increase their model performance without loss data quality. Follow the attributes which will be consider.

  • Flavor: Milk Chocolate, Cherry and Coffe
  • Caloric Content: Sugar
  • Average Package Size: Big and very small

We can not assume a cluster with these attributes together because was made independent analyse. The cluster that has been create represent 97% of the whole volume and 96% of the whole value. The data size decrease by 70% and the error is less than 10% in MAPE, MAE and RSME. Was made “OTHERS” cluster, it is a many small players together that won’t mess up the analysis and keep the chart cleaner. Was applied pareto test to identify the “OTHERS” cluster.

The forecast model used was Prophet developed by Facebbok.

Brand market

Chocolate Milk

The last month result in 2017, the brand Lili drop by 64,8% share volume and Gen up by 25,4% share volume, if you analyse the Price Index will be possible to realise the price has a strong influence because the brand Lili rose the price by 40% over the market price and the Glen brand dramatically dropped by 23% of market price.

The Lily was chosen to apply a forecast, even the last share volume had a suddenly drop this won’t reflect a long term.

Cherry

The Cherry flavor has a scene more equal of share volume by 2015, the next year Harley brand has big variation along the year. In overall, Harley gained share in 2016, but 2017 loss the strength.

Harley has good prediction to 2018.

Coffee

Edshan has almost a stationary graphic. Few big variation but in general keep close to average result line.

Edshan forecast is decreasing along 2018.

Producer market

Chocolate Milk

The Chandler has the same result than your brand Lili. Seems to be the only one Chocolate Milk from Chandler or Lili has the majority then other Chocolate Milk from Chandler.

Cherry

The Cherry market does not have a producer with strong presence, that could be a opportunity.

Coffee

The Chandler has majority share in coffee market.

Contribution

The contribution is a important KPI, because is possible to verify which brand is truly responsible of gain or loss of share.

The cherry contribution of Ali producer has two product cannibalizing each other.

The cherry contribution of Goodman producer need do a deeply analyse. The chart below does not say so much.

The cherry contribution of Chandler producer need do a deeply analyse. The chart below does not say so much.

The coffee contribution of Chandler producer has two product cannibalizing each other.

Conclusio

The report show how the market is performing along months, also which attributes has more relevance in the final result. With these information will be possible filter the data leaving unecessary informaction, also increasing the perfomance for trainning models. In the second part was possible to understand the big brands and how is expected to perfoman in few months forward. The same is for big Producer and the contirbution rate of your brand in the market.

Forward Studies

  • Develop a model with external insight then predict how the market will react.
  • Rewrite on Python for be more flexible.
  • Apply forecast model to others brand and producers.
  • Interactive plot in the report.
  • Month and Year label in x axis share volume.
  • Shiny application in the charts.
  • Erro result of all models.